Accepted Manuscript

Title: Conductivity of Cerium Doped BaFeO3−� andApplications for the Detection of Oxygen

Author: <ce:author id="aut0005" biographyid="vt0005">William D. Penwell<ce:author id="aut0010"biographyid="vt0010"> Javier B. Giorgi

PII: S0925-4005(13)01148-9DOI: http://dx.doi.org/doi:10.1016/j.snb.2013.09.095Reference: SNB 16010

To appear in: Sensors and Actuators B

Received date: 3-5-2013Revised date: 20-9-2013Accepted date: 23-9-2013

Please cite this article as: W.D. Penwell, J.B. Giorgi, Conductivity of Cerium DopedBaFeO3−� and Applications for the Detection of Oxygen, Sensors and Actuators B:Chemical (2013), http://dx.doi.org/10.1016/j.snb.2013.09.095

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

Page 1 of 32

Accep

ted

Man

uscr

ipt

Conductivity of Cerium Doped BaFeO3-δ and Applications for the Detection of Oxygen

William D. Penwell, Javier B. Giorgi*

Centre for Catalysis Research and Innovation, Department of Chemistry, University of Ottawa, 10 Marie Curie Prvt. Ottawa, Ontario, Canada. K1N 6N5.

E-mail: jgiorgi@uottawa.caFax: +1(613)562-5170; Tel: +1(613)562-5800x6037

Corresponding author: Javier B. Giorgi*Centre for Catalysis Research and Innovation, Department of Chemistry, University of Ottawa, 10 Marie Curie Prvt. Ottawa, Ontario, Canada. K1N 6N5.E-mail: jgiorgi@uottawa.caFax: +1(613)562-5170; Tel: +1(613)562-5800x6037

Page 2 of 32

Accep

ted

Man

uscr

ipt

Conductivity of Cerium Doped BaFeO3-δ and Applications for the Detection of Oxygen

William D. Penwell, Javier B. Giorgi*

Centre for Catalysis Research and Innovation, Department of Chemistry, University of Ottawa, Ottawa, Ontario, Canada. K1N 6N5.

E-mail: jgiorgi@uottawa.caFax: +1(613)562-5170; Tel: +1(613)562-5800x6037

Abstract

The effect of Ce doping on the structure and the conductivity of BaFeO3 perovskite

materials is investigated and the resulting materials are applied as oxygen sensors. The new

perovskite family, Ba1-xCexFeO3-δ (x=0, 0.01, 0.03, and 0.05), was prepared via a sol-gel

method. Powder XRD indicates a hexagonal structure for BaFeO3 with a change to a cubic

perovskite upon Cerium doping at the A site. The solubility limit of Ce at the A site was

experimentally determined to be between 5-7 mol %. Bulk, electronic and ionic conductivities of

BaFeO3-δ and Ba0.95Ce0.05FeO3-δ were measured in air at temperatures up to 1000˚C. Cerium

doping increases the conductivity throughout the entire temperature range with a more

pronounced effect at higher temperatures. At 800˚C the conductivity of Ba0.95Ce.05FeO3-δ reaches

3.3 S/cm. Pellets of Ba0.95Ce.05FeO3-δ were tested as gas sensors at 500 and 700˚C and show a

linear, reproducible response to O2.

Keywords: perovskite, gas sensing, conductivity, barium ferrite, Ce doping, oxygen sensor

Page 3 of 32

Accep

ted

Man

uscr

ipt

1. Introduction

Perovskites of the type ABO3 where A=lanthanide and B=transition metal have been

demonstrated for a range of applications such as catalysis, anode/cathode materials for solid

oxide fuel cells (SOFCs) and electrochemical gas sensors [1–4]. Substitution of various cations

into the A and B sites allows for fine tuning of the material properties to better suit the

application. The inclusion of alkaline earth metals at the A site has shown to produce oxides with

high electronic as well as ionic conductivity [5,6]. These mixed electronic and ionic conductors

(MIEC’s) are of current interest as cathode materials for SOFC’s due to their stability and

conductivity in high temperature oxidizing atmospheres.

These parameters are equally important in the field of electrochemical gas sensors, where

p-type semiconductors are used for the detection of oxidizing gases such as O2, NO2 and ozone

[3,7]. In particular, the development of temperature independent O2 sensors is an area of intense

research for the automobile industry. The measurement of pO2 in automobile exhaust systems is

currently performed using expensive zirconia based lambda sensors [8]. To reduce

manufacturing costs, several groups have focused on semiconducting oxides that show resistivity

dependency towards pO2. Materials based on TiO2 and CeO2 have been investigated along with

several perovskite materials that show promising temperature independent O2 sensor behavior

[9–11]. For example, perovskite materials LaCu0.3Fe0.7O3−δ and La0.05Sr0.95Ti0.65Fe0.35O3−δ have

been demonstrated as O2 sensors with low temperature dependency and fast response times [12].

Also, a previous study showed that doping BaFeO3 with Ta increases its sensitivity towards pO2

while maintaining a temperature independent response [13]. Post et al. explored the non-

stoichiometric SrFeO3-x system showing that not only is the equilibrium between the oxide ions

and the gas phase directly related to sensor sensitivity, but also intrinsic phase transitions

Page 4 of 32

Accep

ted

Man

uscr

ipt

resulting from rearrangement and changes in the oxidation state of the iron ions produce drastic

changes in sensitivity at specific oxygen partial pressures and temperatures. These non-

stoichiometric perovskites show extremely high values of sensitivity when tested as

conductometric O2 sensors [14–16]. It is evident in the literature that Fe based perovskites

exhibit many of the ideal material properties for O2 sensing.

Recent studies within our group have shown that doped samarium ferrites have

favourable characteristics such as high conductivity and stability in reducing atmospheres,

making them useful for SOFC anode materials and gas sensors [4,17–19]. The increased

conductivity in the Sm ferrites arises from Cerium doping at the A site, changing the material

from a p to n-type semiconductor [19]. The majority of Fe based perovskites utilize a trivalent A

site cation (lanthanide) leading to a Fe3+ oxide. Less common are Fe4+ oxides that contain a

divalent A site cation, despite several studies demonstrating enhanced properties such as

ferromagnetism and metallic conductivity [20–22]. Combining the stability and conductivity

enhancing properties of A site Cerium doping with an oxygen deficient Fe4+ oxide should

produce a material with high electronic and ionic conductivity.

Here we report on the synthesis, characterization and electrochemical properties of Ba1-

xCexFeO3-δ (x = 0, 0.01, 0.03, and 0.05). Dopant levels greater than 5% were investigated but led

to phase segregation; we estimate the solubility of Ce at the A site to be between 5-7 mol %. The

x = 0.05 sample showed the most promising results and therefore Ba0.95Ce0.05FeO3-δ is the focus

of this work. The following includes a study of the electronic and ionic conductivity as well as

applications as a gas sensor for O2 at various temperatures.

Page 5 of 32

Accep

ted

Man

uscr

ipt

2. Experimental

2.1. Synthesis

Powders of Ba1-xCexFeO3-δ (x=0, 0.01, 0.03, and 0.05) were prepared by decomposition

of citrate precursors via a conventional sol-gel method. Barium nitrate (Ba(NO3)2, AlfaAesar,

99.9%), cerium nitrate hexahydrate (Ce(NO3)3·6H2O, AlfaAesar, 99.5%) and iron nitrate

nonahydrate(Fe(NO3)3·9H2O, AlfaAesar, >98%) were weighed according to the desired

stoichiometry (Ba1-x + Cex : Fe = 1:1) and dissolved in de-ionized water. The metal nitrate

solutions were combined and added to aqueous citric acid monohydrate (AlfaAesar, 99%) so that

the ratio of citric acid to total metal ion was 1:1. The resulting solution was dried at 120°C for 24

h to form a citrate gel that was then ground and calcined at 950°C for 24 h using heating/cooling

ramp rates of 5°C/min.

2.2. Characterization

Powder X-Ray Diffraction (XRD, Philips PW 1830) was used to determine the phase

composition and lattice parameters of the perovskite powders. The instrument used CuKα

radiation (λ=1.54 Å) and the scattering angle, 2θ, was scanned between 13-90° at a rate of

0.02°/s. Panalytical X’PertHighscore Plus software was used for profile refinement and phases

were indexed to the Powder Diffraction File database (JCPDS, 2012).

Scanning electron microscopy (SEM, JEOL JSM-7500F) with an energy dispersive x-ray

spectroscopy (EDS X-Sight) attachment was used to image the surface morphology of Ba1-

xCexFeO3-δ (x=0, 0.05) powders/pellets and provide qualitative elemental composition.

Page 6 of 32

Accep

ted

Man

uscr

ipt

2.3. Total, Ionic and Electronic Conductivity Measurements

The total conductivity (σT) of the materials was measured on circular pellets using the 2

point contact method at temperatures 25-1000°C in air. Pellets (diameter = 6 mm, thickness = 2

mm) of Ba1-xCexFeO3-δ (x=0, 0.05) were made by pressing 2 grams of powder to 15000 lbs and

sintering in air at 1200°C for 4 h with ramp rates of 2°C/min. Platinum mesh (AlfaAesar,

99.9%) was added as a current collector on each face of the circular pellet by adhesion with

platinum paste followed by removal of the organic binder at 800°C for 1h. The test chamber

setup used a quartz tube heated inside a tube furnace as described previously for sensor

experiments [23].

To determine the contributions of ionic and electronic conductivity to the total

conductivity, an Al blocking electrode was used. The Al electrode blocks the flow of O22-

species, therefore σ = σE. The ionic conductivity was inferred from the difference between σT

and σE. To measure σE, two pellets of Ba1-xCexFeO3-δ (x = 0, 0.05) were made as previously

described and a thin film of Aluminum was added between pellets. The “sandwiched” pellets

were heated to 660°C for 4 h with ramp rates of 2°C/min to soften the aluminum and minimize

contact resistance. Platinum mesh was added as a current collector on each side of the combined

pellets and electronic conductivity measurements were recorded at temperatures 25-1000°C in

air. A schematic view of the different conductivity setups is shown in Figure 1. To ensure

reproducibility, multiple trials of each measurement (with and without blocking electrodes) were

performed under both heating and cooling cycles. The 2-point conductivity measurements were

repeated 3 to 6 times for the various samples yielding errors between 2 and 11%, shown with the

appropriate error bars in the relevant figures.

Page 7 of 32

Accep

ted

Man

uscr

ipt

2.4. Sensor Experiments

Sensor experiments were performed by 4 point conductivity measurements in various

atmospheres. To form the sensor, 4 grams of Ba0.95Ce.05FeO3-δ powder was uniaxially pressed at

15000 lbs to form a circular pellet which was subsequently sintered at 1200°C for 4 h with

heating/cooling ramp rates of 2°C/min. The pellet was cut into a rectangular shape with

dimensions 12 6 1 mm and electrical contacts were made by wrapping 4 platinum wires

around the sensor. Platinum paste was applied to ensure electrode contact and the organic binder

was removed by heating to 800°C for 1h. The sensor was contained in a hollow quartz tube

sealed on each end by rubber septa and heated to the desired temperature using a tube furnace.

Mass flow controllers (MFC) allowed for the mixing of gases to create the desired atmosphere

that is then flowed through the quartz tube and over the sensor [23]. Figure 1c shows the 4 point

measurement used for conductivity where the two exterior contacts supplied DC current (100

mA) and the inner two electrodes monitored voltage drop with a digital multimeter. The response

of each sensor was calculated according to Equation 1. Here, the sensor response, S, is defined as

the ratio of the relative conductivity change and allows for the comparison of sensors in various

atmospheres.

[1]

The conductivity of the material was recorded at 1 minute intervals and the flow of test

gas through the test chamber was varied from 2.5-15 sccm with a balance of air so that the total

gas flow rate was 50 sccm.

Page 8 of 32

Accep

ted

Man

uscr

ipt

3. Results and Discussion

3.1. Characterization

The effect of Cerium doping on surface morphology was investigated with scanning

electron microscopy (SEM). Figure 2 a,b shows low magnification images of un-doped BaFeO3-δ

and Ba0.95Ce0.05FeO3-δ respectively. Both samples have an inhomogeneous dispersion of particle

sizes, ranging from less than 500 nm to over 1 µm. Higher magnification imaging reveals a

similar trend in grain boundary sizes (Figure 2 c,d). Similar particle and grain sizes for doped

and undoped materials indicate that Ce doping has little effect on morphology, however the

BaFeO3-δ sample shows signs of phase separation. Higher magnification imaging of BaFeO3-δ

(Figure 2 c) depicts a uniform particle coated in a porous, powder-like material. This could be

indicative of a second phase present in the un-doped sample as is visible through further

characterization with x-ray diffraction.

X-Ray diffraction was used to determine the phase composition of the Ce doped

perovskites. Figure 3 shows the p-XRD spectra for the calcined powders of Ba1-xCexFeO3-δ

(x=0, 0.01, 0.03, 0.05). Spectra are in order of decreasing Cerium content: x = 0.05, 0.03, 0.01,

and 0 corresponding to a, b, c, and d, respectively. The spectrum for BaFeO3-δ (Figure 3 d)

shows a combination of phases resulting from the varying degrees of oxygen vacancies present.

This is a result of the preparation method, namely high temperature calcination in air. The

synthesis of stoichiometric BaFeO3 has been reported by further treatment in pressurized O2

environments and heating in ozone [21]. The prominent phase can be indexed to BaFeO3-δ (PDF:

00-023-1024) in a hexagonal structure with lattice constants a,b = 5.66 and c = 13.83. Other less

Page 9 of 32

Accep

ted

Man

uscr

ipt

prominent phases are visible including a cubic BaFeO3-δ. The multiple phases visible are

consistent with SEM imaging. Doping the material with Ce (Figure 3. a,b,c) leads to a structural

shift from hexagonal to cubic. The Ba0.95Ce0.05FeO3-δ sample (Figure 3. a) shows the cleanest

spectrum with a prominent phase that fits to a cubic perovskite (Pm-3m space group) with lattice

constants a = b = c = 4.035 Å. Rietveld refinement of the powder spectra, along with calculations

of the Goldschmidt tolerance factors, conclude that Ce replaces Ba at the A site in the perovskite

structure. Cerium doping has shifted the material from a hexagonal structure to a cubic

perovskite.

3.2. Conductivity Properties

The total conductivity of each sample was measured in various atmospheres and at

temperature intervals from 25-1000˚C. Both BaFeO3-δ and Ba0.95Ce0.05FeO3-δ show p-type

semiconductor behavior namely higher conductivities in oxidizing than reducing atmospheres.

Tests under a reducing atmosphere (5% H2 in N2, 500°C) led to decomposition of the perovskite

phase and the formation of several phases, including Fe0. This indicates that these materials are

unstable in reducing atmospheres at elevated temperatures. Similar tests performed under air and

O2 indicate complete stability of these materials in oxidizing atmospheres up to 1200°C. The total

conductivities in air from 25-1000˚C are shown in Figure 4 with the appropriate error bars at

each temperature. Cerium doping increased the conductivity of the perovskite throughout the

entire temperature range with a more pronounced effect at higher temperatures. At 800°C the

conductivity of Ba0.95Ce0.05FeO3-δ reaches 3.3 S/cm. This value falls well in the range of other

perovskites containing Ba and Fe [24,25]. Figure 4.b shows an Arrhenius type plot of the

conductivity values between 600-1000˚C and the calculated activation energies. Both samples

exhibit similar changes in conductivity with respect to temperature including a dramatic increase

Page 10 of 32

Accep

ted

Man

uscr

ipt

in ionic conductivity at 800˚C. This is indicative of a change in the dominant conduction type.

Both samples are p-type semiconductors where the prominent charge carriers are electron holes

[2]. As is shown in Equation 2, the replacement of Ba+2 with Ce+3 at the A site in the perovskite

structure decreases the number of oxygen vacancies and increases the number of electron holes.

The greater number of electron holes leads to an increased conductivity for the Cerium doped

material.

[2]

The Barium ferrite perovskite family was expected to be capable of ionic as well as

electronic conduction. It is therefore of interest to determine the individual contributions of ionic

(σI) and electronic conductivity (σE) to the total conductivity (σT) of the material. To measure the

electronic conductivity exclusively, an Al electrode was used as an ionic blocker. This forms a

gradient of oxygen anions on either side of the Al electrode through which only electronic

current can flow. Here it is assumed that σT = σE + σI therefore subtracting the electronic

conductivity from the total conductivity gives an estimate of the ionic conductivity. The

electronic conductivities of BaFeO3-δ and Ba0.95Ce0.05FeO3-δ were measured in the temperature

range of 25-1000˚C (shown in Figure 5.a with the appropriate error bars as measured) and were

subtracted from the bulk values to give the ionic conductivities (Figure 5.b). Both BaFeO3-δ and

Ba0.95Ce0.05FeO3-δ are largely ionic conductors throughout the temperature range. The addition of

5% Ce increased the electronic conductivity by an order of magnitude, from 0.04 to 0.56 S/cm at

800˚C. The ionic conductivity for doped and un-doped BaFeO3-δ increases in a similar manner

until 600˚C at which point the value for Ba0.95Ce0.05FeO3-δ increases whereas BaFeO3-δ decreases.

Page 11 of 32

Accep

ted

Man

uscr

ipt

This indicates that 5% Ce doping has increased both σE and σI. Previous work has attributed the

increased resistance of BaFeO3-δ above 200 K to the Verwey transition (ferromagnetic to

antiferromagnetic transition) [21,26]. However, this explanation has been debated and is beyond

the scope of this work.

3.3. Gas Sensing Properties

The Ba0.95Ce0.05FeO3-δ perovskite has shown relatively high ionic conductivity and

is therefore of interest as a gas sensor, in this case for the detection of oxidizing gases.

Semiconducting gas sensors must have a porous microstructure with high surface area in order to

maximize response[27]. The measured response, conductivity in this case, is directly related to

grain size and separation. Here, sensor experiments were carried out as 4 point conductivity

measurements on a sintered pellet of Ba0.95Ce0.05FeO3-δ. The surface morphology of the sensor

was imaged with SEM (Figure 6). The sensor has a porous microstructure with well defined

grain boundaries and surface structures are apparent with sizes in the 100 nm range.

The sensor signal toward O2 was tested at 500˚C and 700˚C with a carrier gas of air. The

operating temperatures were chosen based on the conductivity measurements previously

discussed. A solid state sensor should aim to maximize sensor response at reduced temperatures.

The Ba0.95Ce0.05FeO3-δ material exhibited a sharp increase in conductivity under oxidizing

atmospheres above 500°C, therefore temperatures of 500°C and higher should result in a greater

sensor response. The sensor maintained a baseline conductivity in air and a range of O2

concentrations were introduced through the test chamber to determine sensor response. The

concentration of O2 through the test chamber is calculated from a base of compressed air

containing 20.95% O2 by volume. The sensor response to a range of O2 concentrations (25-45%)

Page 12 of 32

Accep

ted

Man

uscr

ipt

is shown in Figure 7. The sensor shows a completely reversible response to O2 at both 500˚C and

700˚C. The time required for the sensor to achieve 99% of the stable response values (response

time) and for it to recover its original conductivity upon removal of the target gas (recovery time)

is an important measure in the design of sensors. Response and recovery times for the

Ba0.95Ce0.05FeO3-δ sensor are listed in Table 1. The sensor response to O2 increases with

temperature as expected and the response and recovery times at 700˚C are greater than those at

500˚C. At 500°C it is assumed that the sensor response is dependent upon adsorbed oxygen

species on the material surface [28,29]. This temperature is a typical switchover point below

which not only the cations, but also the oxygen non-stoichiometry can be considered to be frozen

and therefore the sensing mechanism is dominated by surface effects [30–33] As previously

mentioned, the dominant charge carriers for p-type semiconductors are electron holes. Increasing

the concentration of O2 leads to a more oxidative atmosphere where electrons are more readily

removed from the conduction band. This increases the number of electron holes and in turn the

conductivity. This is shown in Equation 3.

[3]

At temperatures higher than 500°C it is known that oxygen equilibration within the p-

type ferrites occurs rapidly [12,28,34]. Therefore at 700°C it is assumed that sensor response is

dependent on bulk transport of oxygen into the perovskite. This sensing mechanism also

decreases the number of oxygen vacancies in the material and increases the conductivity. Here it

is useful to compare the values of S as plotted in Figure 8. It is important to note that because

these measurements were performed with a 4-point setup, the error bars are smaller than the

symbols in the figure. At 500°C, the sensor shows a linear response to O2 concentration until

Page 13 of 32

Accep

ted

Man

uscr

ipt

approximately 40% O2. Beyond this concentration a deviation from linearity is observed. At

700°C the sensor has a greater response due to bulk oxygen equilibration, however the saturation

effect is more pronounced.

Whereas at low temperatures the surface and bulk of the perovskite may not be in

equilibrium, at higher temperatures equilibration results in a change in the number of oxygen

vacancies concurrent with the reduction of Fe from +4 to +3. The fast equilibration observed in

ferrites is partly due to the affinity of Fe ions to achieve the +3 state [14,16,34]. The equilibrated

perovskite stoichiometry for the undoped material can then be written as BaFe+41-xFe+3

xO3-x/2

[32]. The difference in conductivity modes at low vs high temperature results in different

sensitivity of the sensors under each condition as can be derived from Figures 7 and 8. One

would expect this dependency to decrease significantly in the higher temperature region (T �

500°C) where bulk equilibration dominates the sensing mechanism [35]. Figure 9 shows a plot of

log(σ) vs log(pO2) for the Ba0.95Ce0.05FeO3-δ sensor at 500 and 700°C. The slope, M, is a

measurement of sensitivity for the material at each temperature. Although the pO2 range is

limited, the 5% Ce doped material shows an extremely high sensitivity of M=0.38 at T=700°C.

The high pO2 sensitivity and low Ea for conductivity are promising attributes [33]. These

materials may demonstrate temperature independent O2 response at elevated temperatures in the

bulk equilibration range.

4. Conclusion

The Ba1-xCexFeO3-δ (x=0, 0.01, 0.03, and 0.05) perovskite family was prepared via a sol-

gel method. Ce doping at the A site leads to a structural shift from hexagonal to a cubic

Page 14 of 32

Accep

ted

Man

uscr

ipt

perovskite. The incorporation of Ce substantially increases the electronic and ionic conductivity

of these materials, making the perovskites suitable as electrochemical gas sensors. A

Ba0.95Ce0.05FeO3-δ solid state gas sensor exhibited a completely reproducible response to O2 at

500 and 700°C. The sensing mechanism is consistent with a surface process at 500°C and a bulk

process at 700°C.

Acknowledgments

The authors are grateful for financial support from the Natural Sciences and Engineering

Research Council of Canada (NSERC). We would also like to acknowledge the contributions

from the Solid Oxide Fuel Cell Canada Network (SOFCC) and the Center for Catalysis Research

and Innovation (CCRI) at the University of Ottawa.

Page 15 of 32

Accep

ted

Man

uscr

ipt

References

[1] Y. Niu, W. Zhou, J. Sunarso, F. Liang, Z. Zhu, Z. Shao, A single-step synthesized cobalt-free barium ferrites-based composite cathode for intermediate temperature solid oxide fuel cells, Electrochemistry Communications. 13 (2011) 1340–1343.

[2] S.P. Jiang, L. Liu, K.P. Ong, P. Wu, J. Li, J. Pu, Electrical conductivity and performance of doped LaCrO3 perovskite oxides for solid oxide fuel cells, Journal of Power Sources. 176 (2008) 82–89.

[3] Y. Itagaki, M. Mori, Y. Hosoya, H. Aono, Y. Sadaoka, O3 and NO2 sensing properties of SmFe1−xCoxO3 perovskite oxides, Sensors and Actuators B: Chemical. 122 (2007) 315–320.

[4] S.M. Bukhari, J.B. Giorgi, Chemically stable and coke resistant Sm1−xCexFeO3−δ perovskites for low temperature solid oxide fuel cell anode applications, Journal of Power Sources. 198 (2012) 51–58.

[5] J. Matsuno, T. Mizokawa, a. Fujimori, Y. Takeda, S. Kawasaki, M. Takano, Different routes to charge disproportionation in perovskite-type Fe oxides, Physical Review B. 66 (2002) 1–4.

[6] D. Rembelski, J.P. Viricelle, L. Combemale, M. Rieu, Characterization and Comparison of Different Cathode Materials for SC-SOFC: LSM, BSCF, SSC, and LSCF, Fuel Cells. 12 (2012) 256–264.

[7] H. Aono, E. Traversa, M. Sakamoto, Y. Sadaoka, Crystallographic characterization and NO2 gas sensing property of LnFeO3 prepared by thermal decomposition of LnFe hexacyanocomplexes, Ln[Fe(CN)6]·nH2O, Ln = La, Nd, Sm, Gd, and Dy, Sensors and Actuators B: Chemical. 94 (2003) 132–139.

[8] K.H. Friese, F. Stanglmeier, H.M. Wiedenmann, Electrochemical sensor to determine the oxygen concentration in gas mixtures, U.S. Patent WO 95/25276, 1995.

[9] N. Izu, W. Shin, I. Matsubara, N. Murayama, The effects of the particle size and crystallite size on the response time for resistive oxygen gas sensor using cerium oxide thick film, Sensors and Actuators B: Chemical. 94 (2003) 222–227.

[10] A. Takami, Development of titania heated exhaust-gas oxygen sensor, Ceramic Bulletin. (1988) 1956–1960.

[11] R. Moos, W. Menesklou, K.H. Hardtl, Materials for temperature independent resistive oxygen sensors for combustion exhaust gas control, Sensors and Actuators B: Chemical. (2000) 178–183.

Page 16 of 32

Accep

ted

Man

uscr

ipt

[12] K. Sahner, R. Moos, N. Izu, W. Shin, N. Murayama, Response kinetics of temperature-independent resistive oxygen sensor formulations: a comparative study, Sensors and Actuators B: Chemical. 113 (2006) 112–119.

[13] P.T. Moseley, D.E. Williams, Gas sensors based on oxides of early transition metals, Polyhedron. 8 (1989) 1615–1618.

[14] M.L. Post, B.W. Sanders, P. Kennepohl, Thin films of non-stoichiometric perovskites as potential oxygen sensors, Sensors and Actuators B: Chemical. 14 (1993) 13–14.

[15] D. Wang, J.J. Tunney, X. Du, M.L. Post, R. Gauvin, Thermal stability of SrFeO3/Al2O3 thin films: Transmission electron microscopy study and conductometric sensing response, Journal of Applied Physics. 104 (2008) 023530.

[16] J.J. Tunney, M.L. Post, The Electrical Conductance of SrFeO2.5+x Thin Films, Journal of Electroceramics 5:1. (2000) 63–69.

[17] S.M. Bukhari, J.B. Giorgi, Redox Stability of Sm0.95Ce0.05Fe1-xCrxO3-δ Perovskite Materials, Journal of The Electrochemical Society. 158 (2011) H1027.

[18] S.M. Bukhari, J.B. Giorgi, Surface and redox chemistry of Sm0.95Ce0.05Fe1−xNixO3−δ perovskites, Solid State Ionics. 194 (2011) 33–40.

[19] S.M. Bukhari, J.B. Giorgi, Tuneability of Sm(1−x)CexFeO3±λ perovskites: Thermal stability and electrical conductivity, Solid State Ionics. 180 (2009) 198–204.

[20] J.B. MacChesney, J.F. Potter, R.C. Sherwood, H.J. Williams, Oxygen Stoichiometry in the Barium Ferrates; Its Effect on Magnetization and Resistivity, The Journal of Chemical Physics. 43 (1965) 3317.

[21] N. Hayashi, T. Yamamoto, H. Kageyama, M. Nishi, Y. Watanabe, T. Kawakami, et al., BaFeO3: a ferromagnetic iron oxide., Angewandte Chemie (International Ed. in English). 50 (2011) 12547–50.

[22] T. Takeda, R. Kanno, Y. Kawamoto, M. Takano, Metal – semiconductor transition, charge disproportionation, and low-temperature structure of Ca1–xSrxFeO3 synthesized under high-oxygen pressure, Solid State Sciences. 2 (2000) 673–687.

[23] S.M. Bukhari, J.B. Giorgi, Cobalt Doped Sm0.95Ce0.05FeO3-δ for Detection of Reducing Gases, Journal of The Electrochemical Society. 158 (2011) J159.

[24] H. Choi, A. Fuller, J. Davis, C. Wielgus, U.S. Ozkan, Ce-doped strontium cobalt ferrite perovskites as cathode catalysts for solid oxide fuel cells: Effect of dopant concentration, Applied Catalysis B: Environmental. 127 (2012) 336–341.

Page 17 of 32

Accep

ted

Man

uscr

ipt

[25] A.K. Azad, A. Khan, S.-G. Eriksson, J.T.S. Irvine, Synthesis, structure and magnetic properties of Sr2Fe1−xGaxMoO6 (0≤x≤0.6) double perovskites, Materials Research Bulletin. 44 (2009) 2181–2185.

[26] H. Kronmuller, F. Walz, J. Rivas, D. Martinez, Influence of Ba Content on the Magnetic After-Effect Spectra in Barium Ferrites, Physica Status Solidi (A). 137 (1994) 137–148.

[27] N. Barsan, D. Koziej, U. Weimar, Metal oxide-based gas sensor research: How to?, Sensors and Actuators B: Chemical. 121 (2007) 18–35.

[28] W. Gopel, New materials and transducers for chemical sensors, Sensors and Actuators B: Chemical. 19 (1994) 1–21.

[29] H. Iwahara, Ionic Conduction in Perovskite-Type Compounds, in: T. Ishihara (Ed.), Perovskite Oxide for Solid Oxide Fuel Cells, Springer US, Boston, MA, 2009: pp. 45–63.

[30] R. Moos, K.H. Hardtl, Defect Chemistry of Donor-Doped and Undoped Strontium Titanate Ceramics between 1000° and 1400°C, Journal of the American Ceramic Society. 10 (1997) 2549–2562.

[31] B. Boukamp, M. Pham, D. Blank, H. Bouwmeester, Ionic and electronic conductivity in lead–zirconate–titanate (PZT), Solid State Ionics. 170 (2004) 239–254.

[32] A. Rothschild, W. Menesklou, H.L. Tuller, E. Ivers-tiffe, Electronic Structure, Defect Chemistry, and Transport Properties of SrTi1-xFexO3-y Solid Solutions, Chemistry of Materials. (2006) 3651–3659.

[33] P.T. Moseley, Materials selection for semiconductor gas sensors, Sensors and Actuators B: Chemical. 6 (1992) 149–156.

[34] J.J. Tunney, M.L. Post, X. Du, D. Yang, Temperature Dependence and Gas-Sensing Response of Conduction for Mixed Conducting SrFeyCozOx Thin Films, Journal of The Electrochemical Society. 149 (2002) H113.

[35] R. Moos, N. Izu, F. Rettig, S. Reiss, W. Shin, I. Matsubara, Resistive oxygen gas sensors for harsh environments., Sensors (Basel, Switzerland). 11 (2011) 3439–65.

Page 18 of 32

Accep

ted

Man

uscr

ipt

Tables

Table 1. Response and recovery times for Ba0.95Ce0.05FeO3-δ O2 sensor at 500˚C and 700˚C.

Response Time (min) Recovery Time (min)

Peak

Oxygen Content % T= 500°C T = 700°C T = 500°C T = 700°C

A 25 6 9 10 14

B 29 4 17 8 19

C 33 7 17 7 21

D 37 10 14 9 23

E 45 8 10 12 22

Page 19 of 32

Accep

ted

Man

uscr

ipt

Figure Captions

Figure 1. Experimental setups for sensor and conductivity tests showing the 2-point (a.b) and 4-point (c) methods used. The power supply provided a constant current of 100 mA.

Figure 2. SEM images of BaFeO3-δ (a,c) and Ba0.95Ce0.05FeO3-δ (b,d) as-prepared powders.

Figure 3. p-XRD of Ba1-xCexFeO3-δ x = 0-0.05. Spectra are in order of decreasing Cerium content: x = 0.05, 0.03, 0.01, and 0 corresponding to a, b, c, and d respectively. Phase labels: hexagonal = asterisk, cubic = square.

Figure 4. a) Conductivity of Ba1-xCexFeO3-δ (x=0, 0.05) between 25-1000˚C. b) Arrhenius plot of conductivities between 600-1000˚C.

Figure 5.a) Electronic and b) Ionic conductivities of Ba1-xCexFeO3-δ (x=0, 0.05).

Figure 6. Low (a) and high (b) magnification SEM images of Ba0.95Ce0.05FeO3-δ sensor.

Figure 7.Sensor conductivity in O2 at temperatures of a) 700˚C and b) 500˚C. Features are labelled by their O2 concentration: 25, 29, 33, 37 and 45 % O2 for A, B, C, D and E, respectively.

Figure 8. Sensor Response vs. O2 concentration for Ba0.95Ce0.05FeO3-δ sensor at 500˚C and 700˚C.

Figure 9. log(σ) vs log(pO2) plot for Ba0.95Ce0.05FeO3-δ sensor at 500 and 700°C.

Page 20 of 32

Accep

ted

Man

uscr

ipt

Biographies

William D. Penwell received his B.Sc. Honors in Chemistry, in 2011, from Dalhousie University in Halifax, Nova Scotia, Canada. Following a brief stint in industry at Maxxam Analytics, he moved to Ontario, Canada to pursue a M.Sc. at the University of Ottawa. Currently in his second year, his research focuses on the development of perovskite oxide materials for Solid Oxide Fuel Cells (SOFC’s) and electrochemical gas sensors.

Javier B. Giorgi studied chemistry at Concordia University in Montreal. In 1999 he received his Ph.D. from the University of Toronto with a thesis in surface aligned photochemistry. In 2000-2002 he was a Humboldt fellow at the Fritz Haber Institute in Berlin where his work involved the study of model solid catalysts. Since 2002 he has been a Professor in the Department of Chemistry at the University of Ottawa, Canada. His current interests involve single crystal and nano-structured defective oxide materials for applications in solid oxide fuel cell technology, chemical sensors and catalysis.

Page 21 of 32

Accep

ted

Man

uscr

ipt

Figure3(xrd).tif

Page 22 of 32

Accep

ted

Man

uscr

ipt

Figure1.tif

Page 23 of 32

Accep

ted

Man

uscr

ipt

Figure5b.tif

Page 24 of 32

Accep

ted

Man

uscr

ipt

figure6.tif

Page 25 of 32

Accep

ted

Man

uscr

ipt

Figure7a.tif

Page 26 of 32

Accep

ted

Man

uscr

ipt

Figure7b.tif

Page 27 of 32

Accep

ted

Man

uscr

ipt

Figure 2revised.tif

Page 28 of 32

Accep

ted

Man

uscr

ipt

Figure4brevised.tif

Page 29 of 32

Accep

ted

Man

uscr

ipt

Figure(s)

Page 30 of 32

Accep

ted

Man

uscr

ipt

Figure5arevised

Page 31 of 32

Accep

ted

Man

uscr

ipt

Figure8revised

Page 32 of 32

Accep

ted

Man

uscr

ipt

Figure 9